HOME HELP CONTACT US SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Rapid Responses: Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Mulvehill, S. Articles by Porter, A. Search for Related Content PubMed PubMed Citation Articles by Mulvehill, S. Articles by Porter, A.

Template-Guided Versus Undirected Written Medical Documentation: A Prospective, Randomized Trial in a Family Medicine Residency Clinic

From the University of Texas Southwestern Family Medicine Residency Program, the University of Texas Southwestern Medical Center at Dallas

Correspondence: Corresponding author: Sharon Mulvehill, MD, Department of Family Medicine, 6303 Harry Hines Boulevard, Dallas, TX 75390-9165 (e-mail: smulve{at}parknet.pmh.org)

	Abstract

Top Notes Abstract Methods Results Conclusions References

 
Purpose: To compare a template-driven medical documentation system to undirected handwritten documentation and determine whether the template (1) decreases physician evaluation time, (2) increases gross billing, and (3) increases physician satisfaction with the documentation process.

Methods: A prospective randomized trial of documentation with a template system (T-System for Primary Care, Dallas, TX) versus undirected handwritten documentation was conducted in 2 separate teams of a single family medicine residency program. After training, one team used the template system and the other team used undirected written documentation. Clinic visit duration was recorded. Medical records were evaluated by a blinded professional coder to assign an evaluation/management code. Clinic visit duration and coding level differences were evaluated using an independent t test. At the conclusion of the study, residents completed a questionnaire to determine physician satisfaction with the documentation tool. Survey responses were on a –2 to + 2 Likert scale. Means and standard deviations are reported.

Results: A total of 1339 patients were included in the analysis of patient visits. There was no significant difference in clinic time between the template system and the written documentation visits. The mean visit time was 1.75 hours for both teams. For the analysis of gross billing, 1237 charts were included. The mean billing amount for written documentation was $150 and for the template system it was $163—a statistically significant difference. The physicians’ surveys favored continuing to use the template documentation method.

Conclusions: The template medical documentation system compared with undirected written documentation produced a significantly higher bill for the visit, yielding no differences in evaluation time, and was overall positively received by the residents and faculty.

Research is needed to evaluate both EMR and other documentation methods. Results on paper documentation research may offer improvements applicable to many family medicine practices. Lessons learned evaluating paper documentation methods may translate to use on an EMR platform.

The template system chosen for this study is T-System for Primary Care. It is a template-generated documentation system that was first designed for emergency care in 1996. Since then it has been used extensively across the United States. It is currently the documentation method for almost 40% of all emergency department (ED) visits in the United States and exists in both a paper and an integrated EMR format. The paper format was adapted to primary care in 2001 and is currently used by over a thousand providers. The cost of implementation is currently a $500 initial fee and then $2200 per year per provider. To study the effects of the template documentation system, we designed a prospective randomized controlled study. This study hypothesizes that the template system will (1) decrease physician evaluation time, (2) increase coding and reimbursement levels, and (3) improve physician satisfaction with the documentation record.

	Methods

Top Notes Abstract Methods Results Conclusions References

 
The study was performed in an inner city, county-owned, hospital-operated family medicine residency clinic. There are 2 separate teams that draw patients indiscriminately from the same undifferentiated patient population. The 2 teams are in separate but adjacent office space on the same floor of a professional office building. Both teams see equal volume. New patients are assigned to one of the 2 teams based on appointment availability. Thereafter patients are seen by the same team to which they were originally assigned. Residents are assigned to either team at the beginning of their residency and stay in that team through their 3 years of training. A total of 16 residents participated in either the template group or the control group. The clinic treats approximately 10,000 patients per year.

One team was randomly chosen by coin toss to be the study team. The residents assigned to that team received a 1-hour training session on the template system from a company representative. All faculty physicians attended the training. The faculty worked in the same team for the duration of the study with few exceptions. Nurses and medical assistants attended the same training session because they are rotated between teams. The templates were used approximately 6 weeks before the start of data collection.

Three template systems were reviewed before selecting one for use in the study. The selected system was chosen because of higher volume of use and greater customer support. For examples of this template system, please visit http://www.tsystem.com/Template-Systems/primary-care.asp. The template system we used is a series of 40 templates based on chief complaint. Each template includes specific symptoms based on the chief complaint(s). Additional complaints are addressed in a small narrative section. By a series of circles, checks, and backslashes the physician documents the presence or absence of each finding. A tailored review of systems is provided, and there are areas for past medical/social/family history. Relevant physical findings are listed on the template, and they are checked or backslashed to denote the presence or absence. Each of the above sections includes some blank lines for additional information. Shaded areas denote information required for higher levels of service.

There were 2 primary outcomes measured: time of physician evaluation and billing amount. A secondary outcome was physician satisfaction. Physician evaluation time was calculated using the time the patient was placed in the room and the clinic discharge time. A power analysis indicated that a sample size of 650 patients per team had 88% power to detect a 7-minute reduction in time (a clinically relevant number). Based on current clinic volume, a study period of 6 weeks was chosen.

To analyze billing amount, an independent trained and certified coder was hired to determine the billing amounts for all patient visits included in the study. The 1995 Health Care Financing Administration (HCFA) guidelines were used. These guidelines are often used in practice because they are less complex than a 1997 revision. The coder was given no details regarding the nature of the study nor was he shown the actual codes assigned by the family medicine department. Based on the previously published results, a sample size of 650 has a power of 80% to detect a 19% billing increase (an effect size of 0.16).

Both physician evaluation and billing amount were checked for normality and outliers. An independent t test was used to compare both the average examination room time of the 2 teams and the average billing amount across the 2 teams.

At the conclusion of the study a survey was given to the residents of the team using the template to measure physician satisfaction. Permission to use the survey tool was granted by the American College of Emergency Physicians and had been used in a previous study.⁸ Ten questions were administered using a 5-point Likert scale, with responses ranging from strongly disagree to strongly agree. A set of 3 different questions was asked of the faculty physicians using the same Likert scale. The mean response was calculated. Because this is an unvalidated opinion survey, the comparative analysis was considered secondary, and power was not determined a priori.

All analyses and graphs were performed on SPSS 12.0 (SPSS Inc., Chicago, IL). Institutional Review Board permission was granted before onset of data collection.

	Results

Top Notes Abstract Methods Results Conclusions References

 
The control team included 626 patient visits with a mean billing amount of $149.63 (95% CI, 145.8 to 153.5), whereas the template team included 611 patient visits with a mean billing amount of $163.38 (95% CI, 159.0 to 167.7). The gross difference is $13.75 (95% CI, 8.0 to 19.5). An independent t test indicates that the gross billing amount is statistically significant across the 2 teams for the independent blinded standard coder (t = 4.67, df = 1235) (Figure 1). Table 1 shows the frequency for common billing codes by team. The 2 teams saw a total of 1339 patients over the specified time period. Of these, 1237 patient records were sent to the independent coder. That indicates that 102 medical records were lost to coding. Because of blinding, we were unable to determine which patients were missing the coder information, but the loss was equally distributed across the 2 teams.

View larger version (12K): [in this window] [in a new window]   Figure 1. Gross Billing Amount

View larger version (13K): [in this window] [in a new window]   Figure 2. Mean Patient Visit Time

	Conclusions

Top Notes Abstract Methods Results Conclusions References

We chose to measure the time the patient was placed in the room until the patient was discharged to represent the clinic visit time. This time was measurable in a reliable way in our clinic, yet included many variables other than the physician encounter. Our data showed no significant difference in clinic times between the 2 teams. This demonstrates that the template system can be adopted with minimal training time (1 hour) and neither adversely affects nor improves productivity.

The coding levels shown are representative of our experience in an inner city, county hospital clinic (no lower level new patient codes—99201, 99202). There are few simple visits for new patients; they often present with a list of complaints. In addition there were few high level codes for new patients (99205). Higher level of service is more difficult to deliver and document in new patients because documentation guidelines for new patients are more extensive at the higher level of service.

Regarding the coding data, there was significant improvement in overall billing in the team using the template system. The HCFA guidelines are a complex set of directives regarding required documentation for each evaluation and management code. Prior studies have determined that both under and over coding are common errors using these guidelines.⁷,⁸ In addition, expert coders frequently do not agree when assigning a level of service.⁸ This is worrisome because an accusation that a physician is coding improperly can lead to audits and penalties. Complete records with required documentation should yield better billing data by preventing both internal and payer down coding. The mean professional bill was $162 for the template team and $149 for the undirected charting groups, using the 1995 HCFA Guidelines.

Physician satisfaction survey results show that both residents and faculty physicians involved in the study slightly favored using the template system. These results are supported by a prior study that measured satisfaction with the template in the ED 1 month and 1 year after implementation. This prior study found that satisfaction was weakly positive initially and further improved with continued use of the product.⁹ Further studies should address how continued use of the product affects satisfaction in the family medicine setting. In addition, physicians refer back to prior documented visits during patient follow-up. Physician satisfaction with a template system for this purpose should be studied.

The results of our study are consistent with a similar study within an emergency medicine setting.⁸ The Emergency Medicine Study also used the T-System and showed improved billing but a nonstatistically insignificant decrease in patient visit time. Physician satisfaction in the Emergency Medicine Study was higher than in our setting when measured after a longer period of use.

There are several limitations to this study and areas for further study. Both residents and faculty perceived the template documentation as slightly faster. Studying the exact time of physician evaluation rather than total patient visit time might detect a difference between the template system and the undirected format. Speed with the product might improve with continued use so repeating the measurements after longer exposure to the template might yield greater differences.

A further limitation of our study is that it did not index diagnoses or patient complaints. We assumed similar levels of complexity of care because new patients are assigned to either of the 2 teams and subsequently followed in the same team.

This study compared a template-driven medical documentation system to undirected written documentation. Further studies should compare template systems to other types of documentation methods including EMR. Both the residents and faculty are salaried without productivity incentives. There was no financial incentive for them to improve patient billing data. This may affect their satisfaction with the product. Studying a template-driven documentation system in a family medicine setting where billing data affects physician reimbursement might yield improved satisfaction data.

This study looked at billing data but not reimbursement. Further study of charting methods should analyze the actual money generated related to improved billing data. A shift to billing at a higher level requires consideration of additional compliance issues. Medicare patients seen by family medicine residents must also be seen and examined by an attending physician for 99214 and 99215 levels of service.⁹ Local practices may require changes to policy with additional surveillance that may affect the overall generated revenue.

This study focused on billing data, clinic visit time, and physician satisfaction. Further studies on medical documentation should include patient-oriented outcomes. These would include effects on patient satisfaction and on the patient-physician interaction.

The challenge of finding optimal documentation tools will continue in medicine. This study compared a template-driven medical documentation system to undirected written documentation in a family medicine residency clinic. The template system was preferred by the physicians, did not affect clinic times, and improved billing data. On study completion, we chose to implement the use of the template system throughout our clinic.

	Acknowledgments

 
We thank the residents and staff at the Parkland Family Medicine Residency clinic for their support and participation in this study. SM was responsible for study design development, writing, and editing the manuscript. GS and CMC participated in study design development and editing; SR participated in revising and editing, along with providing professional expertise in the area of family medicine. BF was responsible for data analysis and editing, and AP participated in revising and editing and was responsible for study coordination.

	Notes

Top Notes Abstract Methods Results Conclusions References

 
Funding: This study was supported by a grant from T-System for Primary Care (Dallas, TX).

Conflict of interest: none declared.

This study was a poster presentation at the 2004 Scientific Assembly of the American Academy of Family Physicians, 13–17 October 2004, in Orlando, FL.

Received for publication January 25, 2005. Revision received June 8, 2005. Accepted for publication June 13, 2005.

	References

Top Notes Abstract Methods Results Conclusions References

 

1. Wrenn K, Rodewald L, Lumb E, Slovis C. The use of structured, complaint-specific patient encounter forms in the emergency department. Ann Emerg Med 1993; 22: 805–12.[Medline]

2. Schriger DL, Baraff LJ, Rogers WH, Cretin S. Implementation of clinical guidelines using a computer charting system. JAMA 1997; 278: 1585–90.[Abstract]

3. Schmit J. Health care’s paper trail is costly route. USA Today 2004 July 20;Sect. B: 1.

4. Brailer D. The decade of health information technology: delivering consumer-centric and information-rich health care. Available from: http://www.hhs.gov/onchit/index.html.

5. Kikano GE, Goodwin MA, Sange KC. Evaluation and management services. A comparison of medical record documentation with actual billing in community family practice. Arch Fam Med 2000; 9: 68–71.[Abstract/Free Full Text]

6. King MS, Sharp L, Lipsky MS. Accuracy of CPT evaluation and management coding by family physicians. J Am Board Fam Pract 2001; 14: 184–92.[Abstract]

7. King MS, Lipsky MS, Sharp L. Current procedural terminology coding: do the experts agree? J Am Board Fam Pract 2000; 13: 144–6.[Medline]

8. Marill KA, Gaubarou ES, Nelson BK, Peterson MA, Gonzalez MR. Prospective, randomized trial of template-assisted versus undirected written recording of physician records in the emergency department. Ann Emerg Med 1999; 33: 500–9.[Medline]

9. Centers for Medicare & Medicaid Services, Carriers Manual, Fee Schedule for Physician’s Services. Section 15016—Supervising physicians in teaching settings. November 2002. Available from: http://www.cms.hhs.gov/manuals/14_car/3b15000.asp#_15016_0.

This Article Abstract Full Text (PDF) Rapid Responses: Submit a response Alert me when this article is cited Alert me when Rapid Responses are posted Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Mulvehill, S. Articles by Porter, A. Search for Related Content PubMed PubMed Citation Articles by Mulvehill, S. Articles by Porter, A.